As a user I would like a mode that more aggressively predicts author strings

[mjy]

Given the following assumptions are always true:

• A string contains one or more of name, author, and year
• That string contains only those elements (i.e. there is never going to be some element that is not one of those three)

Then I would like to set an option to more agressively predict the author element. False positives are not a major concern, the benefit of good match will be seen downstream via a edit-distance calculation element.

The canonical example is: s = aus Jones.

When I do ::Biodiversity::Parser.parse(s).dig(:authorship, :normalized) then I get parsed: false. I would like to get "Jones".

[mjy]

Thinking further this likely holds for Year too.

[dimus]

hm, parser does require the binomial for species, may be I can have an option to add a 'fake' binomial for zoological species names?